ProCreate, Don't Reproduce! Propulsive Energy Diffusion for Creative Generation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916349335306240 |
|---|---|
| author | Lu, Jack Teehan, Ryan Ren, Mengye |
| author_facet | Lu, Jack Teehan, Ryan Ren, Mengye |
| contents | In this paper, we propose ProCreate, a simple and easy-to-implement method to improve sample diversity and creativity of diffusion-based image generative models and to prevent training data reproduction. ProCreate operates on a set of reference images and actively propels the generated image embedding away from the reference embeddings during the generation process. We propose FSCG-8 (Few-Shot Creative Generation 8), a few-shot creative generation dataset on eight different categories -- encompassing different concepts, styles, and settings -- in which ProCreate achieves the highest sample diversity and fidelity. Furthermore, we show that ProCreate is effective at preventing replicating training data in a large-scale evaluation using training text prompts. Code and FSCG-8 are available at https://github.com/Agentic-Learning-AI-Lab/procreate-diffusion-public. The project page is available at https://procreate-diffusion.github.io. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_02226 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ProCreate, Don't Reproduce! Propulsive Energy Diffusion for Creative Generation Lu, Jack Teehan, Ryan Ren, Mengye Computer Vision and Pattern Recognition In this paper, we propose ProCreate, a simple and easy-to-implement method to improve sample diversity and creativity of diffusion-based image generative models and to prevent training data reproduction. ProCreate operates on a set of reference images and actively propels the generated image embedding away from the reference embeddings during the generation process. We propose FSCG-8 (Few-Shot Creative Generation 8), a few-shot creative generation dataset on eight different categories -- encompassing different concepts, styles, and settings -- in which ProCreate achieves the highest sample diversity and fidelity. Furthermore, we show that ProCreate is effective at preventing replicating training data in a large-scale evaluation using training text prompts. Code and FSCG-8 are available at https://github.com/Agentic-Learning-AI-Lab/procreate-diffusion-public. The project page is available at https://procreate-diffusion.github.io. |
| title | ProCreate, Don't Reproduce! Propulsive Energy Diffusion for Creative Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2408.02226 |